# Quebec Trees Dataset, CanopyRS Preprocessed Tiles: UAV RGB Tree Crowns of a Temperate Forest through Autumn Leaf Phenology

## Facts

- Provider: Cloutier, Germain and Laliberté; preprocessed by the CanopyRS authors
- License: CC-BY-4.0 (https://zenodo.org/records/8148479)
- Size: 16 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/CanopyRS/QuebecTrees
- Access: public, readable without an account
- DOI: https://doi.org/10.5281/zenodo.8148479
- Category: Environmental Science
- Source: https://huggingface.co/datasets/CanopyRS/QuebecTrees
- Page: https://scigantic.com/archive/3b55a0d3-db1e-5f0f-aba4-c3d14f52eb27
- Keywords: Environmental Science, image segmentation, benchmark, Parquet, Hugging Face

## About

A tiled, model-ready version of the Quebec Trees Dataset of Cloutier, Germain and Laliberté, prepared by the CanopyRS authors for the SelvaBox tree crown detection benchmark. The original survey flew a drone (UAV) over a temperate forest in Quebec, Canada on seven dates between May and October 2021, producing high-resolution RGB orthomosaics that follow the same trees through autumn leaf senescence and color change; about 23,000 tree crowns were segmented and labeled in 14 classes. In this release the orthomosaics are cut into tiles with COCO-format crown annotations and split into 148 train, 139 validation and 168 test tiles, stored as Parquet shards under data/ (about 6 GB) and loadable with datasets.load_dataset. Useful for tree crown detection and segmentation, tree species mapping and studying how phenology affects canopy segmentation. The full original deposit, with point clouds and all dates, is on Zenodo (DOI 10.5281/zenodo.8148479). Original data licensed CC BY 4.0; cite Cloutier, Germain and Laliberté (2023) and the SelvaBox paper.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("CanopyRS/QuebecTrees")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/test-00000-of-00003.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/3b55a0d3-db1e-5f0f-aba4-c3d14f52eb27
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/3b55a0d3-db1e-5f0f-aba4-c3d14f52eb27
- Over MCP: `get_schema_card` with id `3b55a0d3-db1e-5f0f-aba4-c3d14f52eb27` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Benchmark datasets](https://scigantic.com/topics/benchmark)
- [Image segmentation datasets](https://scigantic.com/topics/image-segmentation)
- [Environmental Science datasets](https://scigantic.com/topics/environmental-science)
